Representations & Warranties in M&A Agreements: India Guide

Last verified: 30 June 2026

In 2008, a Japanese pharmaceutical major paid roughly US$4.6 billion for a controlling stake in one of India’s best-known generic-drugs companies. It looked like a landmark cross-border deal, the kind that gets written up as a coming-of-age moment for Indian M&A. Then it became something else entirely: a cautionary tale about what happens when the representations and warranties in an M&A agreement turn out to be false. The dispute that followed turned almost entirely on those clauses.

The acquirer later alleged that the Indian target’s former promoters had concealed pending US FDA and US Department of Justice investigations into the target’s manufacturing practices at the time of signing. In plain terms, the buyer said the sellers had breached the representations they gave about the company’s regulatory condition: the promises that the business was compliant, that there were no undisclosed investigations, that what the buyer was paying for was what the buyer thought it was getting.

The sellers contested it. But a Singapore-seated arbitral tribunal sided with the acquirer, awarding roughly Rs 3,500 crore (about US$525 million) in 2016. When the sellers resisted, the buyer brought the award to India for enforcement, and the Delhi High Court upheld it in a judgment delivered on 31 January 2018 (Daiichi Sankyo Company Ltd. v. Malvinder Mohan Singh, O.M.P.(EFA)(COMM) 6/2016). Half a billion dollars changed hands, and the legal hinge of the entire fight was a set of factual statements made in a share purchase agreement years earlier.

That is what a representation is for. It is a contractually allocated, enforceable promise about the target’s condition, and it is the basis on which a buyer claws back value when that promise proves false. Strip away the deal size and the cross-border drama, and the Daiichi-Ranbaxy story is the cleanest possible illustration of why these clauses exist. The buyer could not have known, from the outside, what the sellers knew from the inside. The representations were how the buyer transferred that risk back onto the people who did know.

Here’s the thing most junior lawyers underestimate. Every corporate associate who drafts or reviews a share purchase agreement is drafting the exact category of clause that decided a half-billion-dollar award. The reps block is not boilerplate. It is the part of the contract a litigator will read first if the deal goes wrong, and the part a buyer’s counsel fights hardest over before the deal closes. Master it, and you own the most negotiated, highest-leverage real estate in any M&A document. That single skill is what separates an associate who proofreads schedules from one who runs the risk-allocation table.

So what exactly are representations and warranties, and why do they sit at the centre of every M&A negotiation?

Representations and warranties are factual statements the seller (and, more narrowly, the buyer) makes in an M&A agreement about the target company’s condition. A representation is an assertion of an existing or past fact that induces the other party to enter the deal; a warranty is a contractual promise that the stated fact is, and will remain, true. Together they allocate risk between buyer and seller and form the contractual basis for indemnification if any statement turns out to be false.

That is the topic in a paragraph. The rest of this guide unpacks every working part: the legal framework under Indian law, the types of warranties, the qualifiers and disclosure schedules that shape seller exposure, indemnity mechanics with real India deal-data, sandbagging, W&I insurance, breach remedies, and a practitioner’s walkthrough of how to draft and negotiate the whole package.

What are representations and warranties in an M&A agreement?

Every acquisition runs on a basic asymmetry. The seller has lived inside the target for years and knows where the problems are buried. The buyer, no matter how thorough its diligence, is looking at the company from the outside through a keyhole. So how does a buyer pay good money for a business it cannot fully see? It does not, until the seller commits, in writing and with consequences, to a set of facts about what that business actually is. That commitment is the representations and warranties block.

Representations and warranties are the factual statements a party makes about the target company at signing and, usually, again at closing. The seller represents that the company owns the shares being sold, that its accounts are accurate, that it has paid its taxes, that it is not facing undisclosed litigation, that its key contracts are valid, and so on across a long catalogue. Each statement is a contractual allocation of a specific risk. If the statement is false, the buyer has a remedy, typically an indemnity claim, against the party that made it.

The practical reality is that R&W convert unknowable risk into priced, recoverable risk. A buyer cannot diligence its way to certainty on every employee dispute or tax position inside a target. What it can do is require the seller to stand behind specific facts, so that if a hidden liability surfaces after closing, the loss lands on the seller who knew (or should have known) rather than the buyer who could not have known. To see where this fits, it helps to understand [INTERNAL-LINK: the mechanics of an M&A transaction in India -> M&A in India overview/hub post], because the R&W block is one stage in a longer deal process. That is the entire economic point.

Representation vs warranty: the legal distinction that actually matters

In everyday deal talk, lawyers say “reps and warranties” as if they were one thing. They are not. A representation is an assertion of fact made to induce the other party to enter the contract; if it is false, the wronged party’s classic remedy is rescission (unwinding the deal) plus damages in the nature of restoring the position before the contract. A warranty is a contractual term; if it is breached, the remedy is damages for breach of contract, putting the buyer in the position it would have been in had the warranty been true. Same words on the page, two different legal engines underneath.

Why does this matter in practice? Because the measure of recovery can differ. A misrepresentation claim sounds in the tort-flavoured idea of undoing a deal induced by a false statement; a warranty claim sounds in contract, measuring the gap between the company as warranted and the company as delivered. Indian SPAs almost always label these statements as both “representations and warranties” precisely to keep both routes open, and then narrow the remedies by contract.

Where R&W sit in the deal documents

In a share acquisition, the representations and warranties live in the share purchase agreement (SPA), usually as a dedicated article with the detailed statements pushed into a schedule (the “warranty schedule” or “reps block”). The disclosure schedule sits alongside, carving out exceptions. Where the buyer takes a stake but the parties continue as co-owners, a shareholders’ agreement (SHA) governs the ongoing relationship, and a lighter set of reps may sit there, with the heavy diligence-backed warranties still in the SPA. The architecture matters because a warranty buried in the wrong document can be governed by the wrong limitation regime.

Why R&W are the most heavily negotiated part of an SPA

Ask any transactional lawyer which clauses eat the most negotiation hours, and the answer is reliably the warranties, the disclosure schedule, and the indemnity. The reason is simple: this is where money moves after the deal. Price is agreed early and rarely reopened, but the R&W package decides who pays when something the buyer was promised turns out to be untrue. A common question newer associates raise is whether reps and warranties are “the same thing” and therefore safe to skim. They are not the same thing, and skimming them is exactly how a buyer ends up with a remedy on paper that collapses in practice. Get the distinction right, and the rest of the deal architecture starts to make sense.

Representation vs warranty vs indemnity vs covenant

A buyer’s protection in an M&A agreement is not one mechanism. It is four, working together: representations, warranties, indemnities, and covenants. Drafters bundle them, and casual readers blur them, but each does a distinct job and triggers a distinct remedy. Confuse them and you will either over-promise as a seller or under-protect as a buyer.

Start with the two we have already separated. A representation is a statement of existing or past fact that induced the deal; a warranty is a contractual promise that a stated fact is true. An indemnity is different in kind: it is a promise to reimburse the other party, rupee for rupee, for a specifically identified loss, without the claimant having to prove breach, causation, or remoteness in the ordinary contractual way. A covenant is different again: it is a promise about future conduct, an obligation to do or not do something between signing and closing, or after closing.

Think of it this way. Reps and warranties look backward and at the present (“this is true now”). Covenants look forward (“we will behave like this going ahead”). Indemnities are the payment rail that turns a broken promise into cash. The table below lays out how they line up.

Warranty vs indemnity: which protects the buyer better

For a buyer, an indemnity is usually the stronger remedy, and here is why. To recover on a warranty breach, a buyer must prove the warranty was untrue, that it suffered loss, that the loss flowed from the breach, and it must navigate the common-law duty to mitigate and the rules on remoteness of damage. An indemnity sidesteps most of that: if the identified loss occurs, the seller pays, on a rupee-for-rupee basis, often without the mitigation and remoteness defences applying. That is why buyers push to convert known, specific risks (a pending tax demand, an identified litigation) into specific indemnities rather than leaving them to the general warranties.

What experienced practitioners know is that the warranty/indemnity choice maps to certainty. Unknown, diffuse risk stays in the warranties. Known, identified risk gets pulled out into a bespoke indemnity. A buyer who leaves a flagged diligence problem inside the general warranties, rather than carving it into a specific indemnity, has quietly weakened its own remedy.

Covenants: the temporal difference

Covenants are the future-tense limb of the package. A seller covenants to run the target in the ordinary course between signing and closing, not to declare a surprise dividend, not to dispose of key assets, and to obtain the consents the deal needs. Post-closing, covenants might include non-compete and non-solicit obligations. Breach of a covenant is breach of contract, and because covenants are about conduct rather than a state of facts, the remedy often includes injunctive relief, not just damages. So while a warranty asks “was this true?”, a covenant asks “did you do what you promised?”

Why the four are bundled but do different jobs

The drafting reason they travel together is that a buyer wants overlapping nets. The same defect might be a breach of a warranty (the accounts were inaccurate), trigger an indemnity (the specific tax shortfall), and reflect a covenant breach (the seller failed to disclose a material change before closing). Each net has different proof requirements and different remedies, and a careful buyer wants more than one to fall back on. The drafting discipline is to make sure the nets reinforce rather than contradict each other.

The Indian legal framework: ICA 1872, SOGA 1930 and the uncodified distinction

Here is a fact that surprises a lot of people drafting their first SPA. The crisp representation-versus-warranty distinction that anchors every M&A textbook is not actually written into the core statute that governs the contract. Indian deal lawyers have imported a largely English and American construct and bolted it onto a statutory base that does not formally define either term. Understanding that gap is the difference between drafting with the grain of Indian law and drafting against it.

Does the Indian Contract Act, 1872 define representation and warranty?

The short answer is no. The Indian Contract Act, 1872 (ICA) does not contain a definition of “warranty” as a distinct species of contractual term, nor does it codify the representation/warranty hierarchy that the Sale of Goods regime and English law developed. What the ICA does give you is a framework for free consent, misrepresentation, and fraud. So when an Indian SPA labels a clause a “warranty”, that label draws its meaning from contractual interpretation and imported deal practice, not from a statutory definition. The few Indian authorities that articulate the distinction do so largely in the insurance context. See All India General Insurance Co. Ltd. v. S.P. Maheswari, AIR 1960 Mad 484, which drew the line between a representation and a warranty: as a general rule answers to questions are representations, not warranties, and a warranty’s incorrectness defeats the policy whether or not it is material, while a representation avoids the policy only if false and material (or fraudulent).

Misrepresentation under the ICA and the right to rescind

Where the ICA does heavy lifting is on false statements that induce a contract. Under Sec. 18 of the Indian Contract Act, 1872, “misrepresentation” covers, broadly, a positive assertion not warranted by the information of the person making it, certain breaches of duty, and causing a party to make a mistake about the subject of the agreement. When consent to an agreement is caused by misrepresentation, Sec. 19 makes the contract voidable at the option of the party whose consent was so caused. That is the statutory engine behind a buyer’s right to rescind for a false representation, and it is why Indian SPAs care about whether a false statement is a “representation” (rescission territory) or merely a contractual “warranty” (damages territory).

Silence is its own trap. Mere non-disclosure is not automatically fraud or misrepresentation under Indian law; there must generally be a duty to speak before silence becomes actionable. The Supreme Court made this point in Shri Krishan v. The Kurukshetra University, (1976) 1 SCC 311, holding that mere silence does not by itself amount to fraud, particularly where the other party could have discovered the truth with ordinary diligence. That principle is exactly why disclosure schedules and express anti-sandbagging or pro-sandbagging drafting matter so much: parties cannot safely rely on the background law to fill gaps they left silent.

Does the Sale of Goods Act, 1930 apply to share warranties?

A reasonable question, because the Sale of Goods Act, 1930 (SOGA) does define “warranty” (a stipulation collateral to the main purpose, breach of which gives a right to damages but not a right to reject the goods) and distinguishes it from a “condition”. But shares are generally treated as actionable claims or movable property in the nature of securities, and a share sale is not a straightforward sale of “goods” in the SOGA sense. In practice, Indian deal lawyers do not rely on SOGA to define SPA warranties; the SOGA condition/warranty taxonomy is a useful conceptual analogy, not the governing law of a share-purchase warranty. The warranty’s meaning comes from the contract itself.

The Indian-law approach vs the US/UK approach

How did Indian SPAs end up looking the way they do? Historically, as cross-border M&A into and out of India grew, Indian drafting absorbed the US and UK deal architecture wholesale: baskets, caps, de minimis thresholds, survival tiers, materiality scrapes, sandbagging provisions. English authority treats representations and warranties as legally distinct concepts, a point illustrated in Idemitsu Kosan Co. Ltd. v. Sumitomo Corp., [2016] EWHC 1909 (Comm) (an English Commercial Court decision, persuasive only and not binding in India). The result is a hybrid: an English/American clause vocabulary sitting on an Indian Contract Act foundation that never formally adopted the distinction. Early signals suggest this hybrid is here to stay, with Indian courts increasingly comfortable enforcing the imported constructs through ordinary contract interpretation. For a drafter, the lesson is to define your terms inside the agreement rather than assume the statute supplies them.

What seller representations and warranties typically cover

Once you accept that the warranties carry the contractual weight, the next question is obvious: weight about what, exactly? The catalogue of seller representations is long, and most of it is predictable, but the predictability is the point. A buyer wants a comprehensive map of the company’s condition, stated as facts the seller stands behind. Anything left off that map is risk the buyer silently absorbs.

Who gives R&W: buyer or seller?

Both, but not equally. The seller gives the overwhelming bulk of the representations and warranties, because the seller controls the information about the target. The buyer gives a much shorter set, focused on its own capacity to do the deal. So when people say “the warranties”, they almost always mean the seller’s warranties; the buyer’s are comparatively thin and rarely the subject of serious negotiation.

The standard seller catalogue

A typical seller warranty schedule runs across these categories: title to and ownership of the shares being sold; the seller’s and the target’s capacity and authority to enter the transaction; the accuracy of the financial statements and accounts; tax (filings, payments, no undisclosed demands); litigation and disputes (none pending or threatened beyond what is disclosed); intellectual property (ownership and non-infringement); employment and labour matters; material contracts (valid, in force, no defaults); real property; environmental and regulatory compliance; and the absence of undisclosed liabilities. Each of these is a risk bucket, and each typically becomes its own warranty (or cluster of warranties) tied to the disclosure schedule.

A common question is how granular the catalogue should get. The honest answer is that granularity tracks diligence: the more a buyer’s diligence surfaces a category of risk (say, a target with heavy IP or a complex tax history), the more specific and tightly drafted the warranties in that category become. Generic warranties are the floor, not the ceiling.

Buyer-side representations

The buyer’s side is short by design. The buyer typically represents that it has the corporate capacity and authority to enter and perform the agreement, that it has the funds (or committed financing) to pay the consideration, that no consents or approvals are required that it has not obtained or cannot obtain, and that entering the deal will not breach any law or contract binding on it. The seller cares about exactly one thing here: certainty of payment and completion. That is why the funds and authority representations are the ones a seller’s counsel actually reads.

Types of warranties: fundamental vs business vs tax

Not all warranties are created equal, and treating them as a single undifferentiated block is one of the clearest signs of an inexperienced drafter. Indian SPAs sort warranties into tiers, and the tier a warranty sits in decides how long it survives, whether it is capped at a fraction of the price or the whole price, and how hard it is fought over. Get the tiering wrong and you can accidentally cap a title warranty at the same low level as a routine operational one. So which warranties belong where?

Fundamental warranties and whether they really survive indefinitely

Fundamental warranties go to the root of the deal: title to the shares, the seller’s capacity and authority, and the basic corporate existence of the target. If these are false, the buyer arguably did not get the thing it paid for at all. Because they are existential, fundamental warranties are usually negotiated to survive far longer than ordinary warranties, often for the full statutory limitation period, and sometimes described loosely as surviving “indefinitely”. The practical reality is that nothing survives truly forever: even a fundamental warranty is bounded by the limitation period for bringing a contractual claim. “Indefinite” in a term sheet usually means “as long as the law allows”, not literally without end. They also tend to be capped at up to 100% of the consideration, reflecting that a title failure can unwind the entire economic bargain.

Business (operational) warranties: the diligence-driven core

Business warranties (also called operational or general warranties) are the large middle tier: accounts, contracts, employees, IP, compliance, the working condition of the business. These are the warranties diligence is built to test, and they are the ones most often qualified by materiality and knowledge and most often disclosed against. Because the risk is operational rather than existential, business warranties carry shorter survival periods (commonly around 18 months to 3 years in Indian deals, often tied to one or two audit cycles) and lower caps (a negotiated fraction of the consideration rather than the full price). This is where the basket and de minimis filters do their main work.

Tax warranties and the tax indemnity

Tax sits in its own tier for a reason: tax liabilities can crystallise years after closing, often triggered by an assessment for a pre-closing period. So tax is usually handled through a combination of specific tax warranties and a standalone tax indemnity, with a survival period pinned to the tax statute of limitations rather than the shorter business-warranty clock. In Indian practice, tax survival commonly runs to around seven years, mirroring the window in which tax authorities can reopen assessments. A common question is why tax gets carved out at all: it is because tax exposure is both quantifiable and time-lagged, which makes a dedicated indemnity (rupee-for-rupee, no mitigation argument) a far cleaner tool than a general warranty.

How warranty scope shifts: share deal vs asset deal, SPA vs SHA

The same business throws up different warranties depending on the deal structure. In a share deal, the buyer inherits the company with all its history, so the warranties have to cover the entire corporate body: every past liability, contingent or crystallised. In an asset deal, the buyer cherry-picks defined assets and (usually) leaves the liabilities behind, so the warranties narrow to title and condition of the specific assets and the assumed contracts. That is one reason buyers nervous about hidden liabilities sometimes prefer an asset deal, and sellers prefer a clean share exit.

Document placement matters too. The heavy, diligence-backed warranties live in the SPA, the instrument that effects the transfer. It helps to understand [INTERNAL-LINK: how a share purchase agreement is structured -> Share Purchase Agreement (SPA) guide] before drafting the warranty schedule, because the reps block sits inside that wider architecture. Where the parties remain co-owners, the SHA governs the ongoing relationship and may carry a lighter set of continuing representations. As for drafting them inside an SPA, the warranties are typically stated in the body or a schedule, each one a self-contained factual assertion, and each one read against the disclosure schedule that follows. (We’ll work through the drafting sequence step by step under the practitioner walkthrough.) The table below maps the three tiers against the levers that matter.

Qualifiers and disclosure schedules: materiality, knowledge and the materiality scrape

A warranty as first drafted by the buyer is absolute and unforgiving. The seller’s job in negotiation is to chip away at that absoluteness with qualifiers, and the buyer’s job is to win some of it back with carve-outs to the qualifiers. This back-and-forth is where the real money sits, because a single word like “material” or “knowledge” can shift millions of rupees of exposure between the parties. So how do these levers actually work?

Materiality qualifiers

A materiality qualifier inserts the word “material” (or “in all material respects”) into a warranty, so that only material breaches count. The effect is to filter out trivial inaccuracies: a warranty that “the company is in compliance with all laws” is brutal for a seller, because no company is perfectly compliant; “in compliance with all laws in all material respects” is survivable. For the buyer, the danger is that “material” is undefined and subjective, which is precisely why buyers fight to limit how many warranties get the qualifier and sometimes push back with a materiality scrape (below).

Knowledge qualifiers: actual vs constructive knowledge

A knowledge qualifier limits a warranty to what the seller actually knows: “to the best of the seller’s knowledge, there is no pending litigation.” It shifts the risk of the unknown-unknowns back to the buyer. The decisive drafting question is what “knowledge” means. Actual knowledge is what specified individuals genuinely know; constructive knowledge extends to what they ought reasonably to have known after due enquiry. A seller wants actual knowledge of a small, named group; a buyer wants constructive knowledge of a wide group “after due and careful enquiry”.

Is “to the best of the seller’s knowledge” a loophole? It can be, if left loose. The fix experienced buyers use is to define the knowledge group by name or role, specify whether enquiry is required, and refuse a bare knowledge qualifier on warranties where the seller plainly should know the facts (title, capitalisation, its own filings). What most people miss is that a knowledge qualifier on a fundamental warranty is a red flag: a seller should know whether it owns the shares it is selling, full stop.

The materiality scrape

The materiality scrape is the buyer’s counter-move to materiality qualifiers. It is a clause that reads the word “material” out of the warranties for one or both of two purposes: determining whether a breach occurred, and calculating the loss. A “double materiality scrape” removes materiality for both. The effect is that the seller gets its materiality qualifiers for comfort at the warranty-drafting stage, but when it comes to counting damages, materiality is ignored and the buyer recovers the full quantum. The difference between a materiality qualifier and a materiality scrape is therefore directional: the qualifier protects the seller, the scrape claws that protection back for the buyer at the damages stage.

Disclosure schedules: how they work and where they fail

The disclosure schedule is the document that makes the warranties true. The seller warrants, for instance, that there is no litigation; then it discloses, in the schedule, the three suits that are in fact pending. A disclosed exception is not a breach. There are two disclosure styles: specific (affirmative) disclosure, where each exception is tied to a named warranty, and general (negative or blanket) disclosure, where everything in the data room is deemed disclosed against everything. Buyers want specific disclosure (precise, provable); sellers want general disclosure (a wide safety net).

Where do disclosure schedules fail? Vagueness and incompleteness. A schedule that “discloses” a risk in language too vague to identify it may not protect the seller, and a buyer who accepts a blanket “all data-room contents are disclosed” may forfeit warranty protection it thought it had. Two further mechanics matter: bring-down, where the warranties are repeated at closing and the schedule is updated for events between signing and closing, and the question of whether the seller can update the schedule to cure a breach that arose in that gap. Whether a post-signing schedule update defeats a buyer’s claim is itself a negotiated point, and leaving it silent is how disputes start.

Non-reliance clauses

A non-reliance (or “no-reliance”) clause states that the buyer is relying only on the express warranties in the agreement and not on any other statement, projection, or information provided during diligence. Its purpose is to shut down extra-contractual misrepresentation claims based on stray emails, management presentations, or data-room chatter. For a buyer, the risk is over-narrowing its own remedies; for a seller, it is essential housekeeping to prevent liability for every optimistic statement made during the process. In Indian deals, the enforceability of a non-reliance clause interacts with the ICA’s misrepresentation and fraud provisions, and a clause cannot contract out of liability for fraud.

Indemnification and limitation of liability: baskets, de minimis, caps, escrow

This is the section every seller’s CFO reads twice, because this is where the warranties turn into a number with a ceiling. A buyer that has won a comprehensive warranty package can still walk away with almost nothing if the limitation-of-liability stack is built against it. The stack is a series of filters that a claim must pass through before the seller actually pays, and learning to read it as a system, not a list of clauses, is the single most valuable skill in this whole subject. So what does the stack look like?

How indemnification works

Indemnification is the buyer’s primary post-closing remedy. Rather than suing for breach of warranty in the ordinary way, the buyer makes an indemnity claim under the agreement: it identifies a loss arising from a breach of warranty (or a specific indemnified matter) and the seller reimburses it, subject to the agreed limitations. The same logic that governs [INTERNAL-LINK: how indemnity clauses operate in commercial contracts -> Indemnity clauses in commercial contracts] applies here, sharpened by deal-specific limits. Because indemnity claims are contractual and self-contained, they avoid much of the friction of a common-law damages claim. That convenience is exactly why the limitation-of-liability levers exist: the seller agrees to a clean payment mechanism, but only inside agreed boundaries.

De minimis threshold

The de minimis threshold is the first filter. It is a floor below which an individual claim simply does not count: a claim worth less than, say, Rs 10 lakh cannot be brought at all. Its purpose is to stop the buyer from nickel-and-diming the seller with a flood of tiny claims after closing. Only claims above the de minimis figure are even eligible to enter the basket.

Baskets: tipping vs deductible

The basket is an aggregate threshold: the seller pays nothing until the total of eligible claims crosses a set amount (commonly 0.5% to 2% of the consideration in Indian deals). There are two flavours, and the difference is worth real money. A tipping basket (or “first-dollar” basket): once the aggregate crosses the threshold, the seller pays the whole amount from the first rupee. A deductible basket: the seller pays only the excess above the threshold, like an insurance deductible. So on a Rs 100 crore basket of claims with a Rs 2 crore threshold, a tipping basket pays Rs 100 crore and a deductible basket pays Rs 98 crore. Sellers want deductible; buyers want tipping.

Caps and India deal-data

The cap is the ceiling on the seller’s total liability. In Indian M&A, caps on business warranties commonly range from around 10% to 100% of the consideration, depending on bargaining power, deal type, and whether W&I insurance is in play; fundamental and tax warranties are frequently capped higher (often up to 100%) or carved out from the general cap altogether. The basket and de minimis figures, as noted, typically sit at 0.5% to 2% of consideration for the basket. These ranges come from Indian deal data, and the headline point for a negotiator is that the cap, basket, and de minimis are not independent dials: they are traded against each other across the table. The table below sets the four levers side by side.

Escrow and holdback

A cap and a basket are worthless if the seller has spent the money and disappeared. Escrow and holdback solve the collection problem. In an escrow, a slice of the consideration (commonly 10% to 25%) is parked with a neutral escrow agent for the survival period, available to satisfy indemnity claims; in a holdback, the buyer simply retains part of the price. Either way, the buyer secures a fund it can reach without chasing the seller through litigation. For cross-border deals, the escrow structure also has to work with India’s exchange-control regime, which we come to below.

A worked example with figures

Numbers make the stack concrete. Suppose the consideration is Rs 200 crore. The parties agree a de minimis of Rs 20 lakh per claim, a deductible basket of Rs 3 crore (1.5% of consideration), and a cap of Rs 40 crore (20%) on business warranties. After closing, the buyer surfaces four breaches: Rs 15 lakh, Rs 50 lakh, Rs 2 crore, and Rs 6 crore. The first claim (Rs 15 lakh) is below de minimis and is dropped entirely. The remaining three total Rs 8.5 crore, which clears the Rs 3 crore basket. Because the basket is a deductible, the seller pays only the excess: Rs 8.5 crore minus Rs 3 crore, so Rs 5.5 crore. That figure is well under the Rs 40 crore cap, so the cap does not bite, and the buyer recovers Rs 5.5 crore from (say) the escrow account. Switch the basket to tipping and the seller would instead pay the full Rs 8.5 crore. One word, “tipping” versus “deductible”, moves Rs 3 crore.

Indemnity payments to a non-resident seller: FEMA and tax

Cross-border deals add a layer most domestic-only lawyers forget. When the seller is a non-resident and an indemnity claim succeeds, the mechanics of paying that claim, or of releasing escrowed funds back, engage India’s exchange-control regime under the Foreign Exchange Management Act, 1999 (FEMA) and its pricing and reporting rules, plus withholding-tax considerations. A purchase-price adjustment or indemnity payment flowing across the border is not automatically free to move; it may need to fit FEMA pricing guidelines and reporting, and the tax treatment of an indemnity receipt is itself a contested area. The practical takeaway: a cross-border indemnity clause that ignores FEMA and tax can be unenforceable in substance even if it is perfect on paper.

Sandbagging in Indian M&A

Picture this. A buyer’s diligence team finds a problem: an unpaid tax liability the seller never mentioned. The buyer says nothing, closes the deal, and then, after closing, sues on the warranty that said there were no undisclosed tax liabilities. Can a buyer recover for a breach it knew about before it signed? That question has a name, sandbagging, and in India the answer is unsettled enough that drafters cannot afford to leave it to chance.

What is sandbagging?

Sandbagging is a buyer’s claim on a warranty for a matter the buyer already knew was untrue before closing. The label comes from the idea of holding back, keeping the information in your bag, then deploying it as a post-closing claim. The intuition cuts both ways: a seller feels cheated (“you knew, and you closed anyway”), while a buyer says the warranty is a bargained-for allocation of risk that does not evaporate just because diligence happened to catch the problem.

Pro-sandbagging vs anti-sandbagging clauses

Drafters resolve the uncertainty with an express clause. A pro-sandbagging clause preserves the buyer’s right to claim regardless of its prior knowledge: the warranties stand on their own, and what the buyer knew is irrelevant. An anti-sandbagging clause does the opposite: it bars the buyer from claiming on anything it knew about before closing. Between them lies the most common outcome of all, silence, where the agreement says nothing and the parties are left to whatever the governing law supplies. The difference between a pro- and an anti-sandbagging clause is, bluntly, the difference between a buyer keeping its warranty protection and handing it back.

Is sandbagging allowed in India? The 2023 survey

India has no settled statutory rule, and the market reflects that uncertainty. A 2023 survey of Indian private-target agreements found that roughly 24% were pro-sandbagging, about 6% were anti-sandbagging, and around 70% stayed silent on the issue. Read that again: seven in ten deals leave the single most contested knowledge question unaddressed. The second-order consequence is significant. Because most agreements are silent, the question of what happens lands on courts and tribunals to fill the gap, which means the background judicial trend, not the contract, often decides the outcome in the very deals that did not bother to draft for it.

India’s pro-sandbagging drift and the enforceability question

Which way is the gap-filling drifting? Early signals from Indian deal commentary suggest a pro-buyer (pro-sandbagging) lean: the view that a buyer may rely on the express warranties it bargained for, and that its diligence knowledge does not automatically defeat a contractual claim, drawing some support from the principle that a warranty is an allocation of risk rather than a representation the buyer must have believed. Practitioners expect this drift to continue and to push more parties toward explicit sandbagging drafting. Are anti-sandbagging clauses enforceable in India? In principle yes, as a freely negotiated contractual term, subject to the overriding rule that no clause can shield a party’s own fraud. So which way should a drafter jump? Toward express drafting, every time, because silence in a pro-sandbagging-drifting market is a bet a seller may not want to be making.

Warranty & indemnity (W&I) insurance in the Indian market

There is a quieter way the R&W bargain is being settled in Indian deals, and it does not involve the seller standing behind the warranties at all. Increasingly, an insurer does. Warranty and indemnity insurance has moved from exotic to mainstream on Indian mid-market and private-equity deals over the last few years, and it changes the negotiation in ways that catch out lawyers who learned the subject before it arrived. So what is it, and why is it spreading?

What is W&I insurance and how does it work?

W&I insurance (also called R&W insurance) is a policy that covers loss arising from a breach of the warranties in the SPA. It comes in two forms: a buyer-side policy, where the buyer claims directly against the insurer for a warranty breach, and a seller-side policy, which covers the seller’s liability for warranty claims. Buyer-side policies dominate because they let the buyer recover from a creditworthy insurer rather than chasing the seller. The insurer underwrites the deal by reviewing the buyer’s diligence and the disclosure, prices a premium (a percentage of the cover limit), and sets its own exclusions. In India, the policy interacts with local law and is typically arranged through specialist brokers placing cover with international or domestic insurers.

Why Indian PE/VC sellers now prefer insurance

For a private-equity or venture-capital seller, W&I insurance solves a structural problem: fund life. A fund exiting an investment wants a clean break, returning proceeds to its investors without a contingent indemnity liability hanging over the fund for years after the exit. A buyer-side W&I policy lets the seller give thin warranties (or a token Re 1 indemnity) while the buyer still gets meaningful recovery from the insurer. Early signals suggest this is becoming the default structure on PE-backed Indian exits, precisely because it lets sellers distribute and dissolve without leaving warranty risk on the table. That, more than premium cost, is why insurance is winning.

Tax insurance and what it changes in negotiation

A related product, tax insurance, covers an identified tax risk (for example, the consequences of an uncertain tax position) and has particular pull in India given the country’s history of retrospective and contested tax positions. The second-order effect of all this insurance is a shift in where the negotiation energy goes. When an insurer backs the warranties, the seller’s indemnity exposure shrinks, so the parties stop fighting over caps and baskets and start fighting over policy exclusions, the knowledge scrape, and what diligence the insurer requires. Risk migrates from the seller to the insurer, and with it migrates the lawyer’s job: from negotiating the indemnity to negotiating the policy. The table below sets the three mechanisms side by side.

Breach of representations and warranties: remedies and survival

Everything in this guide converges on one moment: a warranty turns out to be false after the deal has closed. What can the buyer actually do? The answer depends on whether the false statement is treated as a representation or a warranty, how the contract has limited the remedies, and crucially whether the buyer is still inside the survival window. Miss the window and the strongest claim in the world is worthless.

What happens when an R&W is breached

When a warranty is breached, the buyer’s normal route is an indemnity claim (or a contractual damages claim) for the loss, measured as the difference between the value of the company as warranted and its actual value, subject to the basket, cap, and survival limits. When the false statement is treated as a misrepresentation that induced the contract, a different remedy opens up: rescission. The two are not the same. Damages keep the deal alive and compensate the gap; rescission seeks to unwind the transaction altogether. Most SPAs deliberately channel the buyer toward indemnity-style damages and contract out of rescission for ordinary warranty breaches, precisely to keep closed deals closed.

Can a buyer rescind the deal for breach?

Sometimes, but it is the exception. Under Sec. 19 of the Indian Contract Act, 1872, a contract whose consent was caused by misrepresentation or fraud is voidable at the option of the misled party, which is the statutory basis for rescission. But two things usually stand in the way after an M&A closing. First, most SPAs expressly exclude rescission as a remedy for warranty breach, leaving damages or indemnity as the sole route. Second, rescission of a completed share acquisition is often practically impossible: the businesses have integrated, the consideration has moved, and you cannot easily put everyone back where they started. So while the Shri Krishan ruling and the ICA framework keep rescission theoretically available for genuine fraud, in practice the buyer’s real remedy is money, not unwinding. The Daiichi-Ranbaxy outcome, an enforced monetary award rather than a reversal of the share sale, is the realistic template.

Survival periods in India

A survival period is the window after closing during which a warranty remains live and a claim can be brought; once it lapses, the warranty dies and so does the claim. Indian deals tier survival to match risk. Tax warranties (and the tax indemnity) commonly survive around seven years, tracking the period in which tax authorities can reopen assessments. Business or operational warranties typically survive a shorter span, often around 18 months to 3 years, frequently pegged to one or two completed audit cycles so a buyer has time to discover problems. Fundamental warranties (title, capacity, authority) survive longest, frequently for the full statutory limitation period and sometimes described as “indefinite”, though no warranty truly outlives the limitation law. The single most common own-goal is letting a survival clock run out before a known issue has crystallised into a quantified claim.

Does due diligence replace the need for warranties?

A persistent misconception, and a dangerous one. Diligence and warranties do different jobs. Diligence is the buyer’s investigation: it reduces uncertainty and prices risk before signing. Warranties are the contractual backstop for everything diligence could not find, plus a route to recovery for problems that surface later. No diligence exercise is exhaustive; sellers know things buyers cannot reach, and some liabilities (a pre-closing tax position, a latent product defect) only emerge after closing. A buyer who treats thorough diligence as a substitute for warranties has confused investigating the risk with allocating it. You need both: diligence to find what you can, warranties to recover when you could not.

How to draft and negotiate an R&W package: a practitioner walkthrough

Knowing what the pieces are is one thing. Assembling them into a coherent package under deal pressure is another. This is the sequence an experienced transactional lawyer follows, from the first diligence finding to the signed SPA. Each step builds on the last, and skipping one is how gaps open up. Here’s how the package gets built, step by step.

Step 1: Map the risk from diligence to warranties. Start with the diligence findings, not a precedent template. Every material risk the diligence team surfaces should map to either a specific warranty, a specific indemnity, or a price adjustment. Walk the diligence report category by category (title, tax, litigation, IP, employment, contracts) and decide, for each flagged item, whether it becomes a tailored warranty, a carved-out specific indemnity, or a disclosed exception. The output of this step is a risk-to-warranty map, not yet any drafting. This is also where [INTERNAL-LINK: the due-diligence process that feeds the disclosure schedule -> Due diligence in M&A] does its real work: weak diligence produces a weak warranty package.

Step 2: Draft the reps block and tie each warranty to disclosure. Now draft the warranties themselves, each as a clean factual assertion, organised by category and cross-referenced to the disclosure schedule. As you draft each warranty, build the matching disclosure entry, because a warranty without a thought-through disclosure line is either a false statement or a missed protection. Decide upfront whether you are running specific or general disclosure, and keep the schedule precise enough that each exception clearly identifies the matter disclosed. Drafting watertight clauses here is a craft in itself, and it pays to treat [INTERNAL-LINK: drafting watertight contract clauses -> Contract drafting / boilerplate clauses guide] as a discipline rather than a copy-paste exercise.

Step 3: Negotiate the qualifiers and the disclosure schedule. With the warranties drafted, the negotiation moves to qualifiers. Decide which warranties carry a materiality qualifier and which a knowledge qualifier, define the knowledge group by name or role, and settle whether knowledge requires due enquiry. The buyer will press for a materiality scrape on the damages calculation; the seller will resist or limit it. Run this negotiation alongside the disclosure schedule, because every qualifier the seller wins reduces what it needs to disclose, and vice versa.

Step 4: Set the limitation-of-liability stack. Now build the stack: de minimis per claim, basket (tipping or deductible) as an aggregate threshold, cap (tiered by warranty type), and survival periods (tax longest, business shortest, fundamentals longest of all). Negotiate these as one system, trading across the four levers rather than fighting each in isolation. A seller might concede a higher cap in exchange for a deductible basket and a shorter business-warranty survival; a buyer might accept a lower cap in exchange for an uncapped tax indemnity.

Step 5: Decide the security and address sandbagging expressly. Finally, secure the indemnity and close the open questions. Choose between escrow/holdback (a secured fund) and W&I insurance (an insurer-backed alternative), or a combination, and on a cross-border deal, structure the payment mechanics to work with FEMA and tax. Then do the thing most parties avoid: address sandbagging expressly, with a pro- or anti-sandbagging clause, so the agreement does not fall into the 70%-silent gap. A package built through these five steps will survive a real dispute; one assembled from a precedent without this discipline often will not.

Common mistakes and negotiation pitfalls

Even experienced teams repeat the same handful of errors, and each one is a place where a buyer quietly loses protection or a seller quietly takes on exposure it never priced. What are the recurring traps worth flagging?

Vague or under-built disclosure schedules. This is the most common and most expensive mistake. A disclosure schedule drafted in a rush, with vague entries or a lazy blanket “everything in the data room is disclosed”, either fails to protect the seller (because the disclosure is too imprecise to count) or strips the buyer of warranty protection it assumed it had. The disclosure schedule deserves the same drafting care as the warranties it qualifies; treating it as an afterthought is how disputes start.

Leaving sandbagging silent and assuming you are protected. Both sides do this. A buyer assumes it can claim on anything regardless of what it knew; a seller assumes a buyer cannot claim on what it discovered in diligence. In a market where roughly 70% of agreements stay silent and the judicial drift leans pro-buyer, silence does not protect a seller. If it matters to your client, draft for it.

Over-relying on “best of knowledge” qualifiers. Sellers love a knowledge qualifier; buyers should resist it on warranties where the seller plainly should know the facts. A bare “to the best of the seller’s knowledge” on a title or capitalisation warranty is a red flag, and a buyer that accepts knowledge qualifiers across the board has converted a hard warranty package into a soft one without noticing.

Ignoring FEMA and tax on indemnity recovery from non-resident sellers. On cross-border deals, a beautifully drafted indemnity can stall at the payment stage if it ignores India’s exchange-control regime under the Foreign Exchange Management Act, 1999 and the withholding-tax treatment of the payment. The mistake we see most often is a domestic-template indemnity dropped into a cross-border deal without checking whether the money can actually, lawfully, move. Build the FEMA and tax overlay in from the start, not after the dispute.

Frequently asked questions

1. What are representations and warranties in an M&A agreement? They are factual statements a party (mainly the seller) makes about the target company’s condition: its ownership, accounts, tax position, litigation, contracts, and compliance. A representation asserts a present or past fact; a warranty promises that fact is true. Together they allocate risk between buyer and seller and form the basis for an indemnity claim if any statement turns out to be false.

2. What is the difference between a representation and a warranty? A representation is an assertion of fact made to induce the other party to enter the contract; if false, the classic remedy is rescission plus damages. A warranty is a contractual promise that a fact is true; if breached, the remedy is contract damages measured by the gap between the warranted and actual value. Indian SPAs usually label statements as both to keep both routes open.

3. What is the difference between a warranty and an indemnity? A warranty is a promise that a fact is true; recovering on it requires proving breach, loss, causation, and dealing with mitigation and remoteness. An indemnity is a promise to reimburse a specified loss rupee-for-rupee, usually without those hurdles. Buyers prefer indemnities for known, identified risks because recovery is cleaner, and leave diffuse, unknown risk to the general warranties.

4. Which Indian laws govern representations and warranties? The Indian Contract Act, 1872 is the backbone, governing free consent, misrepresentation (Sec. 18) and the right to rescind for misrepresentation or fraud (Sec. 19). The ICA does not formally define “warranty” as a distinct term, so the rep/warranty distinction is imported from English and US practice. The Sale of Goods Act, 1930 defines warranty for goods but does not directly govern share warranties.

5. What are fundamental warranties vs business (operational) warranties? Fundamental warranties go to the root of the deal: title to shares, capacity, and authority. They survive longest (often the full limitation period) and are usually capped near 100% of consideration. Business or operational warranties cover accounts, contracts, employees, IP, and compliance; they survive a shorter window (commonly 18 months to 3 years), carry lower caps, and are most often qualified and disclosed against.

6. What is a disclosure schedule and why does it matter? A disclosure schedule is the document where the seller lists exceptions to its warranties, so a disclosed matter is not a breach. It can be specific (each exception tied to a warranty) or general (blanket data-room disclosure). It matters because vague or incomplete disclosure can either fail to protect the seller or strip the buyer of warranty cover it assumed it had.

7. What is a survival period for representations and warranties? A survival period is the window after closing in which a warranty stays live and a claim can be brought; once it lapses, the warranty and any claim on it die. Survival is tiered by risk: tax around seven years, business warranties roughly 18 months to 3 years, and fundamental warranties for the full limitation period (sometimes loosely called “indefinite”).

8. Can fundamental warranties really survive indefinitely? Not literally. “Indefinite” in a term sheet usually means “as long as the law allows”, because every contractual claim is ultimately bounded by the statutory limitation period. Fundamental warranties (title, capacity, authority) are negotiated to survive far longer than business warranties, often for the full limitation period, but they are not genuinely without end.

9. What is an indemnity cap in an M&A deal? A cap is the ceiling on the seller’s total indemnity liability. In Indian M&A, caps on business warranties commonly range from around 10% to 100% of the consideration, depending on bargaining power and whether W&I insurance is used. Fundamental and tax warranties are frequently capped higher (often up to 100%) or carved out of the general cap entirely.

10. What is a basket (and de minimis threshold) in indemnification? A de minimis threshold is a floor below which an individual claim does not count, filtering out trivial claims. A basket is an aggregate threshold: the seller pays nothing until total eligible claims cross it (commonly 0.5% to 2% of consideration in India). A tipping basket pays everything once crossed; a deductible basket pays only the excess above the threshold.

11. What is the difference between pro-sandbagging and anti-sandbagging clauses? A pro-sandbagging clause lets the buyer claim on a warranty breach regardless of what it knew before closing; the warranty stands on its own. An anti-sandbagging clause bars the buyer from claiming on anything it knew about pre-closing. Where the agreement is silent (about 70% of Indian deals), courts and tribunals fill the gap, and India’s drift leans pro-buyer.

12. What is an escrow or holdback in an M&A deal? Both secure the buyer’s indemnity recovery. In an escrow, a slice of the consideration (commonly 10% to 25%) is parked with a neutral agent for the survival period, available to meet indemnity claims. In a holdback, the buyer simply retains part of the price. Either way, the buyer can reach a secured fund without chasing the seller through litigation.

13. What is W&I (warranty and indemnity) insurance? W&I insurance is a policy covering loss from a breach of the SPA warranties. A buyer-side policy (the common form) lets the buyer claim against a creditworthy insurer instead of the seller; a seller-side policy covers the seller’s liability. The insurer underwrites the deal, prices a premium as a percentage of cover, and sets exclusions. It is increasingly standard on Indian PE-backed deals.

14. Warranty vs indemnity: which protects the buyer better? For known, specific risks an indemnity is usually stronger, because it reimburses an identified loss rupee-for-rupee without the buyer having to prove breach, causation, remoteness, or satisfy the duty to mitigate. A warranty is better suited to unknown, diffuse risk. Sophisticated buyers carve flagged diligence problems into specific indemnities and leave the rest to the general warranties.

15. Is sandbagging allowed in India? There is no settled statutory rule. A 2023 survey found about 24% of Indian private-target agreements were pro-sandbagging, 6% anti-sandbagging, and 70% silent. Where silent, courts fill the gap, and Indian commentary suggests a pro-buyer (pro-sandbagging) drift. Anti-sandbagging clauses are generally enforceable as freely negotiated terms, but no clause can shield a party’s own fraud.

16. What happens if a representation or warranty is breached? The buyer’s usual remedy is an indemnity or contractual damages claim for the resulting loss, measured as the gap between the warranted and actual value, subject to the basket, cap, and survival limits. For a misrepresentation that induced the contract, rescission may be available under the ICA, but most SPAs exclude rescission for warranty breach and channel the buyer toward money damages.

17. Can a buyer rescind the deal for a breach of representation? In theory yes: under Sec. 19 of the Indian Contract Act, 1872, a contract induced by misrepresentation or fraud is voidable at the misled party’s option. In practice, most SPAs exclude rescission for warranty breach, and unwinding a completed acquisition is usually impractical once businesses have integrated and consideration has moved. The realistic remedy after closing is monetary, not reversal of the deal.

18. How long do warranties survive after closing in India? Survival is tiered. Tax warranties and the tax indemnity commonly survive around seven years, matching the tax reassessment window. Business or operational warranties survive a shorter span, often around 18 months to 3 years, frequently tied to one or two audit cycles. Fundamental warranties survive longest, usually for the full statutory limitation period and sometimes described as “indefinite”.

19. What is sandbagging in M&A? Sandbagging is a buyer’s claim on a warranty for a matter it already knew was untrue before closing, then deployed as a post-closing claim. Sellers view it as unfair; buyers view the warranty as a bargained-for risk allocation that does not disappear because diligence caught the issue. Whether it succeeds turns on the agreement’s pro- or anti-sandbagging position, or, where silent, on the governing law.

20. What happens to indemnity claims paid to a non-resident seller (FEMA and tax)? A successful indemnity claim or escrow release involving a non-resident engages India’s exchange-control regime under the Foreign Exchange Management Act, 1999 and may need to fit pricing and reporting rules, plus withholding-tax treatment. A purchase-price adjustment or indemnity flow across the border is not automatically free to move, so a cross-border indemnity that ignores FEMA and tax can fail in substance even if it reads perfectly.

References

Case Law

1. All India General Insurance Co. Ltd. v. S.P. Maheswari, AIR 1960 Mad 484 (Madras High Court; representation vs warranty distinction in insurance law).
2. Daiichi Sankyo Company Ltd. v. Malvinder Mohan Singh, O.M.P.(EFA)(COMM) 6/2016 (Delhi High Court, 31 January 2018; enforcement of the Singapore arbitral award for concealment and misrepresentation at the share sale).
3. Idemitsu Kosan Co. Ltd. v. Sumitomo Corp., [2016] EWHC 1909 (Comm) (English Commercial Court, 27 July 2016; persuasive foreign authority treating representations and warranties as distinct; judgment on BAILII).
4. Shri Krishan v. The Kurukshetra University, (1976) 1 SCC 311 (Supreme Court of India; mere silence is not by itself fraud where the truth was discoverable by ordinary diligence).

Statutes

1. Indian Contract Act, 1872 (sections cited: 18 misrepresentation; 19 voidability and rescission).
2. Sale of Goods Act, 1930 (condition vs warranty distinction; Sec. 12).
3. Foreign Exchange Management Act, 1999 (cross-border indemnity payments, pricing and reporting overlay).

This article is for informational purposes only and does not constitute legal advice. For specific legal guidance, consult a qualified legal professional.